2017 Water Reactor Fuel Performance Meeting September 10 (Sun) ~ 14 (Thu), 2017 Ramada Plaza Jeju Jeju Island, Korea

Transcription

1 COLD SPRAY CR-COATED FUEL CLADDING WITH ENHANCED ACCIDENT TOLERANCE Martin Ševeček 1,2, Anil Gurgen 1, Bren Phillips 1, Yifeng Che 1, Malik Wagih 1, Koroush Shirvan 1 1 Department of Nuclear Science and Engineering, Massachusetts Institute of Technology, Cambridge, MA, USA; sevecek@mit.edu 2 Faculty of Nuclear Sciences and Physical Engineering, Czech Technical University in Prague, Prague, Czech Republic ABSTRACT: Accident Tolerant Fuels (ATF) are currently of high interest to researchers, utilities, the industry as well as governmental and international organizations. One of the widely studied ATF concepts is a multi-layer cladding (also known as coated cladding). This concept is based on a traditional Zr-based alloy (Zircaloy-4, M5, E110, Zirlo etc.) that serves as a substrate. Different ATF materials are then applied on the substrate surface by various techniques; thus enhancing the accident tolerance. This study focuses on results from testing of Zircaloy-4 coated by pure chromium metal using the cold spray technique. In comparison with traditional physical vapor deposition (PVD) methods, the cold spray technique is more cost efficient due to lower energy consumption and more suitable for industry-scale production. Moreover, the cold spray technique does not damage the substrate material or deposition material due to high temperatures. The Cr-coated samples were tested in different conditions (500 C steam, 1200 C steam, PWR pressurization test) and characterized by SEM, EDX or nanoindenter. The results of the steady-state fuel performance simulations showed that such concept is feasible but requires further optimization.. KEYWORDS: Nuclear Fuel, ATF, Coating, Chromium, Cold Spray I. INTRODUCTION Current nuclear fuel systems for light water reactors is based on a combination of a slightly enriched UO 2 pellet and a cladding made of a Zr-based alloy. The system has been used for about a half of a century and its safety and performance has been improved and optimized. However, rare events have occurred where this system does not behave as safely as desired. The accident at the TMI-2 reactor in 1979 and Fukushima Daiichi severe events in 2011 are both examples. Following these events, nuclear industry, utilities, research and governmental organizations started a development program with an objective of development of a nuclear fuel system with enhanced accident tolerance. One of the widely studied concepts is a multi-component cladding (also known as coated cladding). The cladding is considered a near-term technology, which can be developed and employed in several years. The cladding is based on a modification of current Zr-based alloys on which different protective layers are deposited. Many coating materials have been studied around the world as well as different deposition techniques. This paper summarizes results of Zircaloy-4 cladding coated by chromium using cold spray technique. II. EXPERIMENTAL II.A. Material and Sample Preparation Chromium was chosen as a coating mainly due to its extraordinary corrosion resistance, high melting point, good strength, high hardness and high wear resistance. At the same time, chromium does not have a high neutron absorption cross section relative to natural elements such as nickel and is widely available at a reasonable cost. Chromium is expected to be compatible with Zr-based alloys due to similar mechanical and thermal properties, up to their eutectic temperature, which are summarized in TABLE II. Moreover, operating experience with chromium inside reactor cores already exists because some PWR control rods are plated by chromium to increase wear resistance, and chromium is also an alloying element in structural steels. 1

2 II.A.1. Substrate material Standard commercially available annealed Zircaloy-4 was used as a substrate. The composition as measured by EDX of the substrate is summarized in TABLE I. The Zircaloy-4 material was cleaned, descaled and handled before different experiments following the ASTM standard 1. TABLE I: Substrate composition Sn Fe Cr O Zr 1.33± ± ± ±0.01 Balance II.A.2. Cold spray technique TABLE II: Properties of substrate and coating materials at room temperature 2 4 Chromium Zircaloy-4 Ultimate tensile strength [MPa] Yield strength [MPa] Young s modulus [GPa] Elongation [%] 44% 23% Poisson ratio CTE [µm/m- C] Thermal conductivity [W/m-K] Melting point/eutectic [ C] 1860/ /1310 Specific heat capacity [J/g- C] Density [g/cm 3 ] Crystal structure bcc hcp<865 C<bcc Thermal σ a [barn] Substrate materials coated by chromium using various deposition methods have been widely studied 5 7 ; however, the cold spray technique brings new questions and uncertainties to the ATF development process that have to be further investigated. Cold spray technique is based on accelerating particles of the deposition material to very high velocities in a supersonic gas jet. The kinetic energy of particles is converted into plastic deformation after contact of the particle with a substrate causing bonding of the deposition material and the substrate without melting 8 Fig. 1: SEM micrographs of as-coated samples The samples were coated by Air Force Research Laboratory and their as-coated surface is shown in Fig. 1. The quality of cold-spray coatings highly depends on the size of the depositing particles. Relatively large particles (~50-70 µm) 2

3 were used in this case to balance the cost and quality of the coating. Due to the larger particle size the thickness of the coating varied between µm. II.B. 500 C Steam Oxidation Test The choice of the steam testing temperature results from a compromise between the representativeness and duration of the experiment. The uncoated and coated samples were exposed to water steam at 500 o C for more than 20 days. The flow rate varied between 8 11 grams of water/min. According to previous studies, the 500 C steam oxidation should result in material corrosion that is very comparable to typical PWR irradiation assisted corrosion conditions 9. II.B.1. Experimental setup The setup consists of a big tank with deionized water heated by heat plates and a heat tape (steam generator). Steam coming from the steam generator is superheated in a furnace and piped away to the oxidation chamber. The chamber is heated by four independent heat tapes ensuring uniform temperature inside the chamber. Several thermocouples monitor temperature distribution in the system and control the heat up and cooling of the system. The schematics of the setup is shown in Fig. 2. II.C. High-temperature Steam Oxidation II.C.1. Experimental setup Cr-coated closed samples were oxidized in steam at 1200 C. The steam oxidation facility is designed to expose samples to flowing steam in a vertical tube over a range of temperatures ( C), steam flow velocities (1 9m/s) at atmospheric pressure. The schematics of the setup used is shown in Fig. 2. Fig. 2: Experimental setup for high temperature testing (left); Experimental setup for testing in 500 C steam (right) 3

4 The Zircaloy-4 tube was closed by an end plug also made from Zircaloy-4 that was welded by electron beam. The tube was then polished, cleaned, coated with chromium by the cold spray technique and oxidized in three consequent runs. First and second steam exposition was 15 minutes long and last exposition was 60 minutes long. All tests were performed with a flow rate of 5.5 g/min. The sample was quenched in air between each run. Only the first 6-7 cm of the rod is expected to operate at 1200 o C because of the height limitation of the furnace. II.D. Pressurization Test The oxidation facilities are designed to prove that the coating is protective in accident conditions and that it can reduce corrosion and hydrogen pickup during normal operation. One of the main concerns of the multicomponent claddings is adhesion of the coating material during operation. Fuel cladding undergoes various mechanical effects during normal operation including creep, PCMI, etc. Therefore, testing is necessary to determine if the coating survives such conditions to demonstrate it can serve its primary purpose as a protective layer. II.D.1. Experimental setup The pressurization test was designed to show good adhesion properties of cold spray coatings in PWR conditions. The Zircaloy-4 samples were first welded, then coated by chromium and later inserted into a dynamic autoclave, which simulated in-reactor conditions with pressure around 14 MPa and temperature of 310 C. Pure low conductivity water ( µs/m) is used without any additions and ph, inlet and outlet water conductivity, oxygen concentration and electrochemical potential are monitored. The samples are open with atmospheric pressure inside simulating open-gap condition. After 20 days of compression of the cladding, the samples were pressurized by argon up to 38 MPa simulating PCMI conditions for 12 days. Samples were then characterized and the adhesion of the coating was studied. TABLE III: Appearance of uncoated and Cr-coated samples during pressurization test III. RESULTS III.A. 500 C Steam Oxidation Two Cr-coated samples, two pure Cr metal samples, and four Zircaloy-4 samples were oxidized in 500 C up to 20 days. The results of Zircaloy-4 corrosion as shown in Fig. 4 shows very good agreement with published results 9,10. A 4

5 correlation for Zircaloy-4 corrosion was derived and was later used to subtract weight gain from the uncoated parts of the other samples. One side of one of the Cr-coated samples (ST) was used for testing of the fabrication process and the thickness of the coating was considerably lower. The cold-spray coating is very non-uniform in comparison with PVD coatings as can be seen in Fig. 3. However, the quality and also uniformity of the coating can be improved by using finer particles. Even with defects, non-uniform coating and lack of post-fabrication surface treatment, the samples show very high corrosion resistance. The appearance of the samples during the test is summarized in TABLE IV. The pure chromium metal shows extraordinary oxidation resistance and its weight gain is almost negligible. The weight gain of Cr-coated samples is higher due to defects which were found on all coated surfaces. It has to be noted, that the weight gain of Cr-coated samples does not account for surface roughness of the coating which considerably increases the oxidized area. TABLE IV: Weight gains and photos of the sample surface during the test The SEM micrographs in Fig. 3 clearly show different behavior of Cr-coated and non-coated material. The thickness of the Zr-oxide at the edge of the sample reached up to 55 microns whereas Cr-coated surfaces show limited oxidation of the coating as well as the substrate. It can be also seen that Zr-oxide is fragile and susceptible to cracking. The chromium coating remains stable even with low thickness and clearly protects the Zr-based substrate. 5

6 Fig. 3: SEM micrographs of Cr-coated samples after 20 days of oxidation in 500 C. Left: Cr-coated surface (dark) with measured non-uniform thickness; Center: Difference between appearance of coated and uncoated sections of the sample; Right: Thicknesses of Zr-oxide and Cr-coated section of the sample after oxidation. Fig. 4: Weight gains of Cr cold spray coating, Cr pure metal and Zircaloy-4 during oxidation in 500 C steam for up to 20 days III.B. High-temperature steam oxidation The coated sample was oxidized in three consequent experimental runs. During the last run, the sample cracked as can be seen in the Fig. 5. The bottom end cap was made from a bulk Zircaloy-4 material with different crystal orientation in comparison with the tubular section of the sample. Due to anisotropic thermal expansion coefficient of Zircaloy-4, the excessive expansion of the end cap likely caused cracking of the sample. The end cap, as well as different sections of the tubular sample, were later analyzed. 6

7 Fig. 5: Appearance of the Cr-coated sample: 1) Before test, 2) After first 15 minutes test, 3) After second 15 minutes test, 4) After third 60 minutes test Fig. 5 also shows the appearance of the Cr-coated sample before and during oxidation. The appearance of the surface changed during the test similarly to samples oxidized at 500 C (TABLE IV) from as-received silver color over blue and green thin layer of Cr oxide to brown/dark yellow color at the end of the test. SEM micrographs show that tube cracked after approximately 5 minutes of the last exposition. This was revealed based on the thickness of the inner oxide of the sample calculated by Cathart-Pawel correlation 11. The outer Cr-coated surface of the sample proved to be very protective at high-temperature conditions even with very low thicknesses of the coating layer. As it can be seen in Fig. 6, the coating layer thickness ranging between 3 20 µm successfully limited oxidation of the substrate. Fig. 6: Left - tubular part of Cr-coated sample after oxidation at 1200 C; Right - end cap of the sample after high-temperature oxidation 7

8 The inner surface of the sample was exposed to steam for about 55 minutes resulting in the 96 µm thick Zr-oxide. Whereas the outer surface was exposed to minutes of high-temperature steam conditions resulting in oxide which is 2-5 microns thick. Fig. 7: Cr-coated end cap after high-temperature oxidation TABLE V: Atomic composition of the end cap after high-temperature oxidation as measured by EDX using JEOL JSM 6610LS Point #1 Point #2 Point #3 Point #4 Point #5 Point #6 Point #7 Zr [at. %] O [at. %] Cr [at. %] The end cap which caused cracking of the tube showed completely different results in comparison with tubular section of the sample. The chromium coating was still present, but thick oxide layer was found underneath. The Cr-coating was probably damaged at the tip of the end cap and steam oxidized the substrate under the coating causing basically twosided oxidation. Chromium coating was missing in different part of the sample but it has to be pointed out that the sample was very fragile due to the excessive oxidation and some layers could be destroyed during cutting and handling. III.C. Nanoindentation One of the Cr-coated steam oxidation samples was used for a nanoindentation test after oxidation in 500 C steam for 20 days. The nanoindentation sample was prepared by cutting a small strip from the steam oxidation sample of approximately 5 mm x 15 mm. The sample was mounted in epoxy with the cut edge facing upwards and polished with incrementally finer slurries down to 0.3 µm. The sample was then mounted in a NanoTest NTX nanoindenter. An array of 10 x 10 nanoindents on a 10 µm pitch were performed along the Zircaloy-Chromium interface. A Berkovich diamond tip (E=1140 GPa, v=0.07) was used for indentations of 1 µm in depth and loads up to 400 mn. A sample of pure chrome was also cut, mounted, polished, and nano-indented with the same parameters as the Zircaloy-Chromium sample with the exception that an array of only 6 x 6 nanoindents was performed since the sample was homogeneous. 8

9 The resulting average values of the reduced modulus and hardness are summarized in TABLE V. The increase of the hardness at the interface between substrate and deposition material was observed earlier 12, it is caused by the interaction of the sprayed particles with very high kinetic energy with the Zircaloy-4. Interestingly, the cold-sprayed coating after exposure to 500 o C is actually harder than the pure chrome sample. TABLE VI: Results of nanoindentation at the interface oxidized Cr/Zircaloy-4 Reduced modulus [GPa] Hardness [GPa] Zircaloy ± ± 0.48 Interface Cr/Zr ± ± 1.72 Cold sprayed Cr 201 ± ± 0.83 Pure Cr 200 ± ± 0.11 IV. FUEL PERFORMANCE MODELLING Preliminary fuel performance simulations have been performed in finite element based fuel performance code BISON. Material properties like elastic modulus, Poisson s ratio, thermal conductivity, thermal expansion, thermal creep and irradiation hardening for Cr have been implemented into BISON to model the fuel performance of Cr-coating with reference to the traditional uncoated fuel. A simplified PWR power history was used with the constant power of 18 kw/m as well as a standard cosine peaking factor, reaching an average burnup of around 60 MWd/kgU. These simplifications lead to clear comparison in performance of both fuel concepts. Fig. 8: Comparison between fuel performance characteristics of reference uncoated Zircaloy-4 and Cr-coated cladding using BISON code As can be seen from Fig. 8, the Cr/Zr composite behaves very similarly in comparison with reference Zircaloy-4 cladding. The chromium layer is expected to undergo plasticity during normal operation, which has not been experimentally confirmed to date. Plasticity results mainly from different thermal expansion coefficient and swelling strain between the two materials 13. Based on the preliminary calculations, the Cr cold spray concept is promising from the point of expected stress and strains during normal operation of PWRs. However, the simulation is based on available data which are limited to date, and the geometry and other operational parameters were simplified. Therefore, additional simulations and modifications of models are needed in order to precisely predict the behavior of Cr-coated nuclear cladding. 9

10 V. CONCLUSIONS The performance of the cold-sprayed Cr coated Zircaloy cladding was presented. As supported by previous studies using PVD, Cr provides a better high temperature oxidation resistant layer compared to Zircaloy. The coating exhibited good bonding strength and corrosion resistance even though standard commercial grade powder and machinery with no postfabrication surface treatment were utilized. The simulation of the coated Cr cladding under steady state implied small impact on the overall performance of the cladding and feasibility of the concepts. Future work includes further testing, model development and assessment of the coating performance under accident scenarios. ACKNOWLEDGMENTS The support for this work was provided by US Department of Energy Integrated Research Project Grant: DE-NE Chrome Coated samples were provided by Victor Champagne (ARL), Matt Siopis (UTRC) and Aaron Nardi (UTRC). REFERENCES 1. ASTM International, West Conshohocken, PA. ASTM B Standard Practice for Descaling and Cleaning Zirconium and Zirconium Alloy Surfaces. (2016). 2. Weaver, C. W. Irradiation and the ductility of chromium. Scr. Metall. 2, (1968). 3. Stephens, J. R. & Klopp, W. D. High-temperature creep of polycrystalline chromium. J. Common Met. 27, (1972). 4. Holzwarth, U. & Stamm, H. Mechanical and thermomechanical properties of commercially pure chromium and chromium alloys. J. Nucl. Mater. 300, (2002). 5. Brachet, J. C. et al. CEA studies on advanced nuclear fuel claddings for enhanced Accident Tolerant LWRs Fuel (LOCA and beyond LOCA conditions). Contrib Mater Investig Oper Exp LWRs Saf. Perform Reliab Fr Avignon (2014). 6. Kim, H.-G. et al. Adhesion property and high-temperature oxidation behavior of Cr-coated Zircaloy-4 cladding tube prepared by 3D laser coating. J. Nucl. Mater. 465, (2015). 7. Kuprin, А. S. et al. Vacuum-arc chromium-based coatings for protection of zirconium alloys from the hightemperature oxidation in air. J. Nucl. Mater. 465, (2015). 8. Stoltenhoff, T., Kreye, H. & Richter, H. J. An analysis of the cold spray process and its coatings. J. Therm. Spray Technol. 11, (2002). 9. Duriez, C., Dupont, T., Schmet, B. & Enoch, F. Zircaloy-4 and M5 high temperature oxidation and nitriding in air. J. Nucl. Mater. 380, (2008). 10. Leistikow, S. & Schanz, G. Oxidation kinetics and related phenomena of zircaloy-4 fuel cladding exposed to high temperature steam and hydrogen-steam mixtures under PWR accident conditions. Nucl. Eng. Des. 103, (1987). 11. Pawel, R. E., Cathcart, J. V. & McKee, R. A. The Kinetics of Oxidation of Zircaloy 4 in Steam at High Temperatures. J. Electrochem. Soc. 126, (1979). 12. Zou, Y., Goldbaum, D., Szpunar, J. A. & Yue, S. Microstructure and nanohardness of cold-sprayed coatings: Electron backscattered diffraction and nanoindentation studies. Scr. Mater. 62, (2010). 13. Wagih, M., Che, Y. & Shirvan, K. Fuel Performance of Multi-Layered Zirconium Based Accident Tolerant Fuel Cladding. in ICAPP (2017). 10